Does a VPN Slow Down Your Internet?
The short answer is yes — all VPNs introduce some speed overhead. The question is how much. A well-optimized VPN on a fast connection might reduce speeds by 5–15%. A poorly optimized VPN, a distant server, or an overloaded provider can reduce speeds by 50–80% or more. The difference between these outcomes is determined by protocol selection, server choice, VPN provider infrastructure, and your base internet speed.
Understanding VPN speed testing — how to measure it accurately, what the numbers mean, and how to compare providers objectively — lets you make an informed choice and get the best performance from whatever VPN you use.
Why VPNs Slow Down Your Connection
VPNs introduce speed overhead through several mechanisms:
Encryption Processing
All VPN traffic is encrypted and decrypted in real time. This requires CPU processing on both your device and the VPN server. On modern devices with hardware AES acceleration, this overhead is minimal — often less than 5% for AES-256-GCM. On older devices without hardware acceleration, encryption can be a significant bottleneck.
Protocol Overhead
VPN protocols add headers and metadata to each data packet, increasing the total data volume. Additionally, some protocols (like OpenVPN over TCP) add extra acknowledgment overhead that reduces throughput compared to raw connections.
Server Distance and Routing
When you connect through a VPN, your traffic routes from your device to the VPN server, then to the destination. This adds at minimum the round-trip distance to the VPN server in latency. Connecting to a VPN server on another continent can add 80–150 ms of latency and significantly reduce throughput due to the longer path.
Server Load
VPN servers have finite capacity. Popular servers with many simultaneous users experience congestion that reduces available bandwidth. Quality VPN providers monitor server load and expand capacity accordingly. Smaller providers or free VPNs often have severely overloaded servers.
How to Test VPN Speed Accurately
Accurate VPN speed testing requires a controlled methodology. Follow this process:
Establish a baseline. Disconnect from your VPN. Run SpeedIQ's speed test 3 times and average the results. Record your baseline download speed, upload speed, and ping. Test at the same time of day you will run the VPN tests.
Connect to the nearest VPN server. Start with the server geographically closest to you. Run SpeedIQ's speed test 3 times and average the results.
Calculate speed retention. Divide your VPN speed by your baseline speed: (VPN speed ÷ baseline speed) × 100 = speed retention percentage. A good VPN should retain 85%+ of your baseline speed on a nearby server.
Test additional server locations. Test servers in other countries or regions that you plan to use. Note that speed will decrease with distance — this is normal and expected.
Test different protocols. Most VPN clients let you switch between protocols (WireGuard, OpenVPN, IKEv2). Test each and compare results.
Test at different times of day. VPN server performance varies with load. Test at peak hours (evening) and off-peak hours to understand the range of performance.
VPN Protocol Speed Comparison
The VPN protocol has a major impact on performance:
WireGuard
The fastest modern VPN protocol. WireGuard uses a lean codebase (approximately 4,000 lines vs. OpenVPN's 70,000+), modern cryptography (ChaCha20, Poly1305, Curve25519), and is implemented in the Linux kernel for maximum efficiency. WireGuard consistently outperforms other protocols in benchmarks and is now supported by most major VPN providers. Use WireGuard whenever available.
Typical speed retention: 90–95% on nearby servers.
IKEv2/IPSec
Fast, stable, and particularly good at reconnecting after network changes (useful on mobile). Slightly slower than WireGuard but significantly faster than OpenVPN in most implementations. A solid second choice if WireGuard is unavailable.
Typical speed retention: 80–90% on nearby servers.
OpenVPN (UDP)
The most mature and widely audited VPN protocol. OpenVPN over UDP is reasonably fast — not as fast as WireGuard but adequate for most uses. Highly configurable and very secure. Use UDP mode rather than TCP for better performance.
Typical speed retention: 70–85% on nearby servers.
OpenVPN (TCP)
Slower than UDP due to TCP's acknowledgment mechanism. Use only when UDP is blocked (some networks restrict UDP traffic). Not recommended as a primary protocol for performance-sensitive use.
Typical speed retention: 60–75% on nearby servers.
L2TP/IPSec
Older protocol, slower than modern alternatives, and has had known vulnerabilities attributed to it. Avoid unless no other option is available.
Proprietary Protocols
Some providers have developed proprietary protocols: ExpressVPN's Lightway, NordVPN's NordLynx (based on WireGuard), Hotspot Shield's Catapult Hydra. These are often competitive with WireGuard and sometimes faster in specific conditions. Treat them as equivalent to WireGuard in your testing.
What Is a Good VPN Speed?
Speed expectations depend on your use case:
Streaming SD video (Netflix, YouTube): Requires 5 Mbps. Any functional VPN connection should handle this.
Streaming HD video (1080p): Requires 15–25 Mbps. Most good VPNs on nearby servers can maintain this easily.
Streaming 4K video: Requires 25–35 Mbps consistently. Test specifically for sustained throughput, not just peak speeds.
Online gaming: The key metric is latency (ping), not download speed. A VPN adds latency — look for a VPN with a server close to your game's server location to minimize the added delay.
Large file downloads: You want maximum throughput. WireGuard with a nearby server on a fast base connection should deliver 500+ Mbps on gigabit internet.
Video conferencing: Requires both upload and download. Look for consistent 5–10 Mbps upload and low jitter.
Comparing VPN Providers: What to Measure
When comparing VPN providers, measure these metrics consistently:
Speed retention (%): VPN speed as a percentage of baseline. The primary performance metric.
Latency added (ms): Difference between ping with VPN and ping without. Lower is better.
Jitter: Variation in latency. High jitter causes choppy video calls even if average ping is acceptable.
Upload speed retention: Often neglected, but important for video calls, backups, and content creation.
Speed consistency: Run tests at multiple times of day. Consistent performance is more valuable than high peaks with low troughs.
Tips for Maximizing VPN Speed
Use WireGuard: Switch your VPN protocol to WireGuard if available. The difference compared to OpenVPN can be 20–30% faster.
Choose the nearest server: For best performance, connect to the server geographically closest to you. Use a server in another country only when you specifically need that location.
Avoid free VPNs: Free VPNs are almost universally slow due to heavily overloaded servers and limited infrastructure. They also raise serious privacy concerns.
Connect via Ethernet: A wired connection to your router eliminates WiFi overhead and provides more stable speeds for VPN connections.
Close background applications: Applications consuming bandwidth in the background (cloud sync, OS updates, streaming) reduce available bandwidth for your VPN connection.
Enable split tunneling: Split tunneling routes only specific traffic through the VPN, keeping high-bandwidth activities like streaming on your direct connection. This dramatically improves overall performance.
Try a different server: If a specific server is slow, try another in the same region. Server load varies significantly.
Frequently Asked Questions
Can a VPN make my internet faster?
In rare cases, yes. Some ISPs throttle specific types of traffic — streaming video, peer-to-peer downloads — and a VPN can bypass this throttling by hiding the traffic type. In these cases, a VPN may actually improve speeds for throttled activities. For most users and most traffic types, however, a VPN will reduce speed.
Why does my VPN show 100 Mbps but pages load slowly?
Speed and latency are different metrics. A 100 Mbps connection with 200 ms latency will feel slower for browsing than a 50 Mbps connection with 20 ms latency. Browsing involves many small requests, each adding the full round-trip latency. Check your VPN's latency, not just its throughput.
Does a VPN affect gaming ping?
Yes, always. A VPN adds at minimum the latency to the VPN server. If your game server is in Frankfurt and your VPN server is in London, you add London-Frankfurt latency on top of your existing ping. The only exception: if your current route to the game server is inefficient and the VPN server routes through a faster path, the VPN might reduce your total ping. This is rare but documented.
How do I know if my VPN is throttling my connection?
If your VPN speed is consistently far below your baseline speed on a nearby server using WireGuard, the VPN provider may be artificially limiting speeds, or their infrastructure may be overloaded. Test multiple servers and protocols. If all show similar throttling, consider switching providers.
Summary
VPN speed testing requires a baseline, a consistent methodology, and testing across different servers and protocols. WireGuard delivers the best performance in most scenarios. Server proximity is the most important factor for latency; server load and provider infrastructure quality determine throughput.
Use SpeedIQ's speed test to establish your baseline and compare your VPN performance accurately. Good VPNs should retain 85%+ of your speed on nearby servers using modern protocols. Anything significantly below that warrants switching protocols, servers, or providers.